The primary determinants of low bone mass in the elderly and the attendant osteoporotic fractures are bone mass at maturity and subsequent rates of bone loss. In this project, we will evaluate certain variables which we feel are important in the development of peak bone mass and/or the subsequent rates of bone loss, with the hope of identifying high-risk subjects when prophylactic therapy can be most helpful. Project 1 focuses on changes in estrogen/androgen metabolism around the menopause as a cause of rapid bone loss. Since bone loss may begin before menopause, especially from the spine, a group of premenopausal women will also be evaluated. Project 2 is aimed at studying genetic contributions to the determination of peak bone mass and subsequent bone loss and possible genetic-environmental interactions. The twin model will be used for these studies, and the groups to be evaluated include young twins of known placental type (to evaluate prenatal environment influences), and older twins of both sexes (to evaluate differences between the two sexes in genetic expression), and differences which may occur in trabecular and cortical bone. Project 3 will explore the concept that Type II diabetes mellitus and obesity contribute in an additive fashion to slowing of bone loss in postmenopausal women, and that the mechanism of this effect is through the production of estrone from its precursors. Project 4 is a placebo-controlled clinical trial of the use of calcium supplements to increase bone mass in young, growing children. The twin model will be utilized for this trial because of its proven efficiency. These projects will be supported by two Cores. Core A will provide biostatistical and epidemiological support for all of the projects. Core B will provide sex steroid measurements which are critical to Projects 1 and 3.